Tuberculosis infection is a leading cause of death worldwide. The emergence of multi-drug resistant strains of Mycobacterium tuberculosis (Mtb) and the increased prevalence of immunocompromised individuals create the conditions for an explosive epidemic of tuberculosis. Recognition of the limitations in the effectiveness of the M. bovis (BCG) vaccine make developing an improved Mtb vaccine imperative. One new approach to a Mtb vaccine is genetic immunization (aka. DNA immunization). This technique involves directly inoculating simple plasmids encoding a protein of the pathogen into the host cells of the living animal. Since its first demonstration in 1992, it remains a promising vaccine delivery system against a wide variety of pathogens. However, as with the more classical protein vaccines, the problem of identifying the appropriate subunit, in this case DNA, for a vaccine remains. We have developed Expression library immunization (ELI) as a solution to this problem. It is a mechanism for t& rapid screening for immuno-protective epitopes that combines recombinant DNA technology with in vivo infection models. Expression libraries containing the total genome of a pathogen are created and introduced into naive mice. The DNA-inoculated mice are then challenged with the pathogen and tested for increased resistance compared to control animals which have received DNA-libraries from an irrelevant organism. Resistant mice identify libraries that contain plasmids encoding for protective antigenic epitopes. The libraries are divided and the procedure repeated until individual protective plasmids are identified. It is an unbiased, systematic approach for isolating vaccine candidates which can be used as genetic vaccines or in other forms. We have proven the feasibility of this technique with Mycoplasma pulmonis and now propose to use it to develop candidates for an effective Mtb vaccine. We have already initiated this screening process. Specifically, we will: 1. Continue to apply ELI to Mtb using a mouse pulmonary infection model to isolate up to 10 protective genes from the genome. 2. Screen these 10 genes in the guinea pig Mtb model to find the 5 most protective. 3. 0ptimize the presentation of these genes and characterize their immune response/toxicity, performance relative to their corresponding protein products. The ELI approach may provide a new and rapid method for developing a Mtb vaccine as well as for other emerging pathogens.